A compact hub for architecture maps, interactive flow views, API documentation, and the developer reference.
Interactive deep dive into the tool-calling pipeline for walkthroughs and architecture review.
SVGFocused view of the MongoDB hybrid-search agent loop with query rewriting, RRF, and grounding.
Interactive HTMLBrowse the main platform flows and operational lifecycles through a single interactive page.
Long-form documentation for workflows, runtime behavior, modules, and integration patterns.
Swagger UIExplore workflow and evaluation APIs with executable request examples.
SVGInfrastructure and service topology across runtime, storage, security, and third-party systems.
Direct schema files for client generation, endpoint discovery, and API inspection.
We test retrieval with a fixed catalog of 25 questions against a 74-document knowledge base: fictional company documents in English and German, plus 60 decoy documents that reuse the same vocabulary. The run uses real embeddings, real Atlas vector search, and the exact pipeline this app runs in production. No LLM grades anything. A question counts as answered when the passage containing the answer shows up in the results. Last run: August 30, 2026.
| Question type | Questions | Answer in top 5 | Answer in top 10 | MRR |
|---|---|---|---|---|
| One clear fact | 4 | 100% | 100% | 1.00 |
| Fact buried mid-document | 3 | 100% | 100% | 1.00 |
| Exact identifier lookup | 2 | 100% | 100% | 1.00 |
| Paraphrased question | 3 | 67% | 67% | 0.70 |
| German question | 4 | 100% | 100% | 1.00 |
| Answer spans 2 documents, single search | 4 | 75% | 88% | 1.00 |
| Answer spans 3 documents, single search | 1 | 100% | 100% | 1.00 |
| Fact in a linked appendix | 2 | 100% | 100% | 1.00 |
| Multi-document chains, agent follow-up search | 5 | 5 of 5 resolved, every hop ranked first | ||
MRR is 1.00 when the right passage is always the very first result.
Single facts are basically solved. Whether the fact is prominent, buried in the middle of a document, hidden behind an exact ID like NX-4471-KRANICH, or asked in German, the right passage comes back first.
Questions whose answer spans multiple documents behave differently, and the table shows it honestly. One search reliably finds the first document, but the second one only reaches the top 10 for four of five chains. This is why agents search more than once: with a follow-up query, all five chains resolve with the right passage ranked first, including one that crosses three documents.
Two catalog questions have no answer in the knowledge base at all. Both came back with only floor-level scores, so the retriever does not fake an answer when there is none.
A full search takes 333 ms on average (p95: 537 ms): about 170 ms embedding the query, 75 ms in the database, and 90 ms of post-processing. A query costs 14 embedding tokens on average (measured from the API's usage field), and a search returns about 2,300 tokens of context to the model. Re-running the entire benchmark, ingest included, costs around 21,000 embedding tokens, well under a cent, and the freshly inserted chunks were searchable in the vector index after 5 seconds.
The whole benchmark, corpus, questions and runner, lives in the
repository under tools/retrievalBenchmark, so the
numbers can be reproduced at any time.